Recently, Reddit announced that they will be charging for access to their API. While this might bring in money in the short term, they are harming important parts of their eco-system, and some app developers won’t be able to afford API access due to high user numbers. These third party apps are vital to Reddit’s success and turning them away is not a good idea, in my opinion. I think this presents a golden opportunity for someone to replace Reddit in the social media arena by making a similar offering with a free-to-use API.

Currently a well-known startup accelerator is hosting an online startup school. For this, large numbers of people come together via video conferencing, but the tools they are using don’t seem to be up to the job as well as they should be. First there was to be a “founder social” online event, where similar founders were paired via video for 7 minutes at a time, with new connections after each session, for an entire hour.

I recently came across a weird bit of news: Berlin’s ruling mayor Franziska Giffey received a call from Kyiv’s mayor Vitali Klitschko. But then she noticed that this was a fake video call. The counterpart was not the mayor but a synthesized image. She noticed this only after 15 minutes of conversation! This is both one of the more notable deepfake incidents and also a dangerous omen for the future. If a politician can be tricked about the identity of a colleague for this long, anything is possible!

Information workers are often faced with this problem: copy and paste many text snippets from many positions in unstructured or semi-structured text files. This is usually too much work to do manually, while there is no easy way to do this automatically. Current solutions include using ancient Unix tools like grep or coding custom tools just for the one job. This idea is for developing a GUI application which allows users to input sample text positions across multiple files and which generalizes these to the entire collection, going by hints such as common text patterns around the section in question.

Some time ago, most programming was pretty simple: a program was compiled and run on a single computer, at most interacting with others through the network. A lot of today’s programming languages stem from those days. With the advent of the cloud it became possible to run software in many ways and automatically. In the platform-as-a-service (PaaS) model, a finished program is handed to the cloud operator and executed automatically. But this is done with the same old programming model of yesteryear.

There are now more open-source projects on the Internet than you can shake a stick at! My go-to search engine for them is still Google, but that is no longer a good option. So I think there should be a dedicated search engine for open-source software projects and similarly for software products, such as SaaS offerings and closed-source libraries. There are two main challenges for this to work as a startup project.